Noise and Pavement Rehabilitation Using Chip Seal Surfaces Paul R. Donavan 1 Illingworth & Rodkin, Inc. 837 North 11 th Place Ridgefield, Washington 98642 Carrie J. Janello 2 Illingworth & Rodkin, Inc. 429 East Cotati Avenue Cotati, California 94931

ABSTRACT In the United States, increasing numbers of local and state highway agencies are using chip seals for highway pavement rehabilitation due to their relatively low cost and simplicity of application. These surfaces can significantly increase both exterior traffic noise levels and interior noise for vehicle occupants. Recent studies have quantified these effects using pass-by and interior noise and on-board sound intensity (OBSI) measurements. In order to minimize these effects, the California Department of Transportation (Caltrans) has conducted research on different chip seal pavement designs near the City of Bishop, California. Previous measurements of chip seal pavement have indicated an average OBSI level of 106.2 dBA with a range of ±2 dB. The five quietest test sections at Bishop produced an average level of 101.8 dB with a range of ±0.6 dB. In this paper, previous pass-by and interior noise and OBSI measurements are provided along with the comparison of OBSI data to that measured at the Bishop sites. The performance and design information on the Bishop sites are also provided.

1. INTRODUCTION

Chip seal treatments are used as a method of preserving existing pavement and extending life by 5 to 7 years. 1 They are particularly effective for low volume roads such as local and collector roads. However, they also have the potential for windshield damage from loose chips and do not provide structural improvement to the underlying pavement. 2 They are also known to have the potential to produce excessive noise. This has been verified by noise measurements as well as complaints from people residing in proximity to roads that received these treatments. Chip seal pavements consist of a thin layer of bituminous binding agent (or emulsion) placed on existing pavement and then applying and rolling a layer of aggregate on it. Double layer chip seal is an application of single chip seal with the first layer constructed with aggregate one sieve size larger than the second layer. 3 To address highway noise and other concerns, a California Department of Transportation (Caltrans) pavement engineer in Caltrans District 9 had a variety of different chip seal designs installed on fifteen sections of US 395 and California State Route 89 in the Owens Valley east of the Sierra

1 pdonavan@illingworthrodkin.com 2 cjanello@illingworthrodkin.com

Nevada Mountain range. For comparison, one section of conventional rubber asphalt pavement and four sections of Superpave pavement were also installed. To quantify the noise generation of these pavements, Caltrans Headquarters had Illingworth & Rodkin, Inc. (I&R) perform on-board sound intensity (OBSI) measurements on these sections. In addition to a California state-wide OBSI database, I&R also had chip seal OBSI results for pavements in Indiana and Washington State. For the sites in these states, interior and pass-by noise were also available from previous measurement programs.

2. NOISE MEASUREMENT METHODS

2.1. On-Board Sound Intensity All the measurements were made in accordance with AASHTO T 360-16, Standard Method of Test for Measurement of Tire/Pavement Noise Using the On-Board Sound Intensity Method using a new Standard Reference Test Tire (SRTT). Using the dual-probe sound-intensity fixture at a test speed of 97 kph and a ‘cold’ tire inflation pressure of 207 kPa. Air temperatures over the District 9 measurement period were in the range of 11 to 22°C. No temperature corrections were used. The microphone signals were analyzed into 1/3 octave band levels using a 5-second averaging time. The microphones were calibrated using a Larson Davis Model CAL200 acoustic calibrator set for 94 dB at the beginning and end of the measurement period. Three or more passes were made for each test section, which were averaged together during post analysis.

2.2. Additional Noise Measurements Pass-by noise measurements of individual test vehicles were performed following the data acquisition methods defined in the AASHTO Standard Method of Test, TP 98-13, Determining the Influence of Road Surfaces on Vehicle Noise Using the Statistical Isolated Pass-By (SIP) Method. The measurement microphone was placed 7.5m from the centerline of vehicle travel lane, and the test vehicles were operated one at a time passed the microphone at a speed of 97 kph. The sound level of the entire pass-by was acquired, and maximum A-weighted sound pressure level was reported. Interior noise was measured with a microphone position above the right front passenger seat 74cm above the seat cushion with the seat adjusted to its center of fore/aft travel and centered on the headrest. The reported level was the average of 0.1 second levels measured over a 10-second period with the vehicle operating at 97 kph. The test vehicles were measured with their original equipment tires. 3. NOISE ISSUES WITH CHIP SEAL PAVEMENTS

3.1 E xterior Tire-Pavement Noise The application of sound intensity to measuring tire-pavement noise was developed at General Motors (GM) in the early 1980s for research into tire noise source mechanisms. 4 It was later applied at GM to new car tire selection for both pass-by regulatory concerns 5 and interior noise. In the early 2000’s, it was used by Caltrans for evaluating pavements for purposes of traffic noise reduction. As part of the Caltrans research, chip seal pavements were quickly identified as producing high noise levels relative to most other pavements, as shown in Figure 1. 6 This has also been reported in other data sets. OBSI measurements were also made at an automotive test track with pavements designed to replicate those in California. The results are shown in Figure 2 and also indicate how chip seal pavement produces high noise compared to other common pavements. 7 High OBSI levels for chip seal pavements have also been measured at other locations, as shown in Figure 3 in comparison to conventional dense grade asphalt concrete (DGAC) measured in Indiana, Michigan, Washington, and

Figure 1: Overall OBSI levels for various pavement types in California and Arizona

Figure 2: Overall OBSI levels of various pavement types at the Hyundai-Kia test track, Mojave, CA California. In the extreme, the loudest chip seal is more than 10 dB higher than the 9.5mm aggregate DGAC pavement. One-third octave band level comparisons for these roads are shown in Figure 4 along with 19mm DGAC. A one-third octave band comparison of chip seal and DGAC pavements measured in Indiana and Michigan are shown in Figure 5. The largest differences in these pavements occur in the low frequencies, below 1,000 Hz where differences of up to 12.7 dB are observed.

lun

Figure 3: Overall OBSI levels of chip seal and dense graded asphalt pavements in the US

Figure 4: 1/3 Octave band OBSI levels of chip seal and dense graded asphalt in the US 3.2 Pass-by and Interior Noise The higher OBSI levels for chip seal surfaces translate into higher pass-by noise levels. From the Indiana and Michigan measurements, two chip seal and two DGAC pavements were measured. Specification for the chip seal pavements were not available; however, photographs of the four surfaces are shown in Figure 6. Pass-by spectra for these four pavements are shown in Figure 7, as measured for a compact Nissan Versa sedan test vehicle with its original equipment tires. The trends between the interior and OBSI results of Figure 7 and 5, respectively, are similar at 800 Hz and below, with the chip seal pavements producing higher levels compared to the DGAC surfaces

by 5 to 10 dB. At 2,000 Hz and above, both the pass-by and OBSI levels on the chip seal are slightly higher than the DGAC. Similarly, the pass-by noise levels for a Chevrolet Equinox SUV test vehicle displayed elevated levels in the lower frequencies, as shown in Figure 8. Overall pass- by noise increased by 6.4 and 6.6 dBA for the two test vehicles.

Figure 5: 1/3 octave band spectra of chip seal and DGAC pavements measured in Indiana

8 ‘Sound intensity Level, dBA, zuee Z “SN S 1/8 Octave Band Center Frequency, He -t-chip Seal - 107348 -+-Chip Seal #2107548, aanpleanesan —eAaphak #21089 8A

Figure 6: Photographs chip seal (upper left and right) and DGAC (lower left and right) pavement measured in Indiana and Michigan

Figure 7: 1/3 octave band pass-by noise levels on Indiana chip seal and DGAC – small sedan

e 8 Pass-by Sound Pressure Level, dA Bas se SPOS IPS SSI SPEESSPSESSE 1/3 Octave Band Center Frequency, Me pope te irrtny Fo tietephhry

Figure 8: 1/3 octave band pass-by noise levels on Indiana chip seal and DGAC – midsize SUV Chip seal surfaces can produce even larger differences in interior noise. Compared to pass-by noise, interior noise in the lower frequencies below 400 Hz are particularly exaggerated, creating what is often termed as “road roar.” The interior noise produced on the chip seals in Indiana is shown in Figure 9 for a Mazda CX-5 compact SUV. The noise in these lower frequencies is primarily structure- borne and more difficult for vehicle manufacturers to reduce. The overall noise level difference between Chip Seal #2 and Asphalt #2 was 7.0 dB, with the bulk of the difference coming from frequency bands from 100 to 400 Hz. A Toyota Camry mid-size sedan was also measured in some

SPOOL IPI SPP PPPS ES IPED ‘Us octave band center Frequent PchpSeatet «864th —e-ChipSealR2-8L 608K —e-Asphat 1-75.28

of the Indiana chip seal testing and on the chip seal pavements in Washington State. These data, shown in Figure 10, displayed an even greater difference in interior noise level relative to DGAC. This vehicle appears to be particularly susceptible to higher frequency inputs around 1,000 Hz as well as lower frequencies.

Interior Sound Pressure Level, dBA sub 8 8 8 8 a 8 PHP PP OPSPISIS PSPSPS SLSOS 118 octave and center frequency Me -t-WA Chip Seal-70.4 40, “IN chip Sea H2- 70.0484 Asphalt #2 -62.4 dBA

Figure 9: 1/3 octave band interior noise levels on Indiana chip seal and DGAC – midsize SUV

Figure 10: 1/3 octave band pass-by noise levels on Indiana chip seal and DGAC – midsize sedan

PIP OP EOL IP EP LES PESII SSI LS 1/3 octave band center Frequency He -t-Cnip Seale -72.048A, “+-ChipSealn2- 72.8484, ce-Asphalt #1 67.2 d8A “e-Asphalt #2 - 65:8 484

4. RESULTS OF CALTRANS DISTRICT 9 CHIP SEAL OBSI MEASUREMENTS

A total of 20 test sections were measured in District 9 including 15 chip seal test sections with chip sizes from 6.35 to 9.53mm. Of the 15 sections, four were double layer chip seal with the first layer containing 9.53mm chip sizes. For the second layer of the double layer chip seals, the emulsion used was Polymer Modified Emulsion (PME), and chip sizes were 6.35 or 7.94mm. For the first layer, the emulsions were PME, Asphalt Rejuvenating Emulsion (AR), or AR GP 50. There was one section of single layer of 7.94mm chips using PME and fabric over an unspecified pavement. The remaining single layer chip seals were all constructed with 9.53mm chip sizes. The emulsions were either AR or PME. Four sections were of Superpave DGAC with 9.53mm thickness. The final chip seal section was chips on asphalt rubber binder. For reference, the last section was Gap Graded Rubberized Hot Mix Asphalt (RHMA-G). The overall A-weighted OBSI levels for each of the 20 test sections are shown in Table 1 along US Highway 395 or SR 89 at the corresponding post mile (PM) and direction of travel, southbound (SB) or northbound (NB). The spectra for the 6.35mm and 7.94mm chip seals are provided in

Table 1: Test Section Details and Overall A-Weighed Sound Intensity Levels

Section Location 1 Pavement Type OBSI Level,

dBA 1 US 395 PM 108.5, SB-1 Chip Seal: ¼” PME, 3/8” AR 102.2 2 US 395 PM 104.5, NB-1 Chip Seal: 5/16” PME, 3/8” PME 102.5 3 US 395 PM 108.5, NB-2 Chip Seal: 5/16” PME with Fabric 101.6 4 US 395 PM 88.5, SB-1 Chip Seal: ¼” PME, 3/8” AR GP50 101.3 5 US 395 PM 88.5, SB-2 Chip Seal: ¼” PME, 3/8” AR GP50 101.6 6 US 395 PM 49.25, NB-2 Chip Seal: 3/8” PME 104.1 7 US 395 PM 44.1, NB-1 Chip Seal: 3/8” AR 104.4 8 US 395 PM 44.1, NB-2 Chip Seal: 3/8” AR 104.0 9 US 395 PM 29.1, SB-1 Chip Seal: 3/8” AR 104.7 10 US 395 PM 29.1, SB-2 Chip Seal: 3/8” AR 104.3 11 US 395 PM 119.0, NB-2 Gap Graded ½” RHMA-G 96.7 12 US 395 PM 10.0, NB-2 Chip Seal: 3/8” AR 103.2 13 US 395 PM 9.8, SB-2 Chip Seal: 3/8” PME 102.5 14 US 395 PM 10.0, NB-1 Chip Seal: 3/8” AR 103.3 15 US 395 PM 9.8, SB-1 Chip Seal: 3/8” PME 102.9 16 US 395 PM 15.3, SB-2 DGAC: 3/8” Superpave 103.3 17 US 395 PM 15.0, NB-1 DGAC: 3/8” Superpave 100.5 18 US 395 PM 15.3, SB-1 DGAC: 3/8” Superpave 102.8 19 US 395 PM 15.0, NB-2 DGAC: 3/8” Superpave 102.8 20 SR 89 PM 2.7, SB-1 3/8” AC Rubber Chip Steel Wheel Roller 104.2 1 US = United States Highway, SR = California State Route, SB = Southbound, NB = Northbound, 1 indicates inner lane, 2 indicates outer lane Seal

Figure 11. It would be expected that the pavement with the smallest chip size would produce the lowest levels; however, from Figure 11 and Table 1, this trend is not necessarily the case. The two higher levels are indicated by both the 6.35mm and 7.94mm chip seals, and the lower levels also contain these two sizes of chips. The spectra for the 9.74mm chip seals are shown in Figure 12. Although there is some variation in both the spectra and levels of Table 1, the spectra do all appear to have a similar shape with less variation than the 6.35 and 7.94mm chip sizes of Figure 11. For comparison, the spectra for the Superpave test sections are shown in Figure 13. In this figure, three of the Superpave pavements are similar; however, the levels for Section 17 are about 5 dB lower than the other sections in the bands below 1,000 Hz. A visual comparison of the pavement of Sections 17

Figure 11: 1/3 OBSI spectra for 6.35 and 7.94mm chip seal pavements

: = 173 Oxoe Bad ete he

Figure 12: 1/3 OBSI spectra 9.53mm chip seal pavements

R 8 8 8 ‘0851 Sound intensity Level, dBA 1/3 Octave Band Center Frequency, He iene

S WS a 1/3 Octave Band Center Frequency, He “0-38 sperpve Section 16) #3) Superpove (ection 17) Pi irchernameerpanicantacny aa caaeaeat

Figure 13: 1/3 OBSI spectra Superpave pavements and 16 shown in Figure 14 indicates that although both pavements are reported to be 9.74mm, the aggregate for Section 16 appears to have more surface roughness.

Figure 14: Photographs of Superpave pavements – left Section 17, right Section 16 The three groupings of 6.35/7.94mm chip seal, 9.53mm chip seal, and Superpave were averaged and compared in Figure 15 to the 12.7mm RHMA-G. Compared to the chip seal surfaces, the RHMA-G surface has inherently less positive texture as can be seen by visual comparison in Figure 16. As a result, the RHMA-G pavement produces lower OBSI levels throughout the entire spectrum compared to the chip seal surfaces. Figure 15 also indicates that the average of the 6.35/7.95mm surfaces produce levels that are equal to the Superpave. It also illustrates the difference in tire noise level between the 6.35/7.95mm and 9.53mm chip seals for the frequencies below 1,000 Hz due to the greater positive texture of the 9.53mm surface.

& 8 8 (0851 Sound intensity Level, dBA 8 1/3 octave Band Canter Frequency, ME 9635/7 tr average 101.6488) 4-$.Siam teperene Aoomget0.3 000) 4-9 Samm Average (103748)

Figure 15: 1/3 OBSI comparison of averages of the 6.35/7.94mm chip seal surfaces, the average of 9.53mm chip seal surfaces, the average of Superpave surfaces, and the 12.5mm RHMA-G

Figure 16: Photographs of RHMA-G and 6.35mm chip seal pavements 5. CONCLUSIONS

Chip seal pavements are inherently noisier than most other pavements, excluding portland cement concrete pavements with non-transverse tining. This increased noise is apparent in interior vehicle noise, pass-by noise, as well as in OBSI measurements made very near the tire noise source. By controlling the aggregate size to be 9.53mm, chip seal pavements can be at the lower limit of about 104 dBA (OBSI), as documented in other measurements (Figure 3). With the use of chip sizes from 6.35 to 7.94mm, achieving levels in the range of 101 to 102 dBA is attainable and comparable to the upper range of other asphalt surfaces. For chips sizes greater than 9.53mm, levels up to 108 dBA may be expected.

6. ACKNOWLEDGEMENTS

Design and build of the twenty test sections chip seal and other pavements was completed under the direction of John Fox of Caltrans District 9. Funding for the measurements and reporting of the noise levels was provided by Bruce Rymer of Caltrans Headquarters. REFERENCES

1. HMA Chip Seals - Materials & Road Research – MnDOT, 2022 Minnesota Department of Transportation, 395 John Ireland Blvd, St. Paul, MN http://www.dot.state.mn.us/mnroad/projects/hma-chip-seals/index.html 2. Jalali1, F. & Vargas-Nordcbeck, Life-Extending Benefit of Chip Sealing for Pavement Preservation, Journal of the Transportation Research Record, National Academies of Science, Transportation Research Board 2021. 3. Schuler, S., Lord, A., Epps-Martin, A, and Hoyt, D., Manual for Emulsion-Based Chip Seals for Pavement Preservation, NCHRP Report 680, Transportation Research Board, National Academy of Sciences, Washington, D.C., 2011. 4. Donavan, P. and Oswald L., “Quantification of Noise Mechanisms of Blank, Rib, and Cross-Bar Tread Bias-Ply Truck Tires”, General Motors Research Laboratories Publication GMR-3750, Warren, MI, July 1981 5. Donavan, P.R., “Tire-Pavement Interaction Noise Measurement under Vehicle Operating Conditions of Cruise and Acceleration”, SAE Paper 931276, Society of Automotive Engineers Noise and Vibration Conference Proceedings, Traverse City, MI, May 1993. 6. Donavan, P., and Rymer, B., “Quantification of Tire/Pavement Noise: Application of the Sound Intensity Method”, Proceedings of Inter-Noise 2004, Prague, the Czech Republic, August 2004 7. P. Donavan and D. Lodico, “Tire/Pavement Noise of Test Track Surfaces Designed to Replicate California Highways”, Proceedings of Inter-Noise 2009, Ottawa, Ontario, Canada, August 2009.